Ryujinx Git (2020/01/10) is complied. Ryujinx is a Experimental Switch emulator written in C#. Don't expect much from this. Some homebrew apps works, and Tetris shows the intro logos (sometimes) but that's about it for now. Contributions are always welcome. Ryujinx Git changelog:
* InitializeCounters to Initialize
* PrintGpuInformation
* Add GPU Information to Logging
* Fix R5G6B5 in SurfaceFlinger (#863)
* Add runtime identifiers to new projects
* Address PR feedback
* Remove some unused args on the shader translator
* Fix FADD32I check for Ra negation
* Remove FFMPEG dependency from README
* Address PR feedback
* Fix some shader disposal issues
* Fix some spelling mistakes
* Fix ShaderTools
* Address PR feedback
* Target .NET Core 3.0 for NVDEC project
* Re-add NVDEC project (not integrated)
* More code cleanup
* Remove more unused code
* GPU resource disposal
* Add XML documentation to Ryujinx.Graphics.Gpu
* Add XML documentation to Ryujinx.Graphics.Gpu.Engine
* Add XML documentation to Ryujinx.Graphics.Gpu.State
* Add XML documentation to Ryujinx.Graphics.Gpu.Shader
* Add XML documentation to Ryujinx.Graphics.Gpu.Memory
* Change GPU errors to debug and reduce log messages
* Fix typos pointed out by LDj3SNuD
* Remove some error logging to avoid log spam
* Fix missing S8 format on texture compatibility check
* Add XML documentation to Ryujinx.Graphics.Gpu.Image
* Fix AstcDecoder Success property never being set to true
* Some code cleanup
* Add basic error logging to the GPU
* ASTC optimizations (#845)
* Reimplement limited bindless textures support
* Initialize GPU physical memory accessor from KProcess, to allow homebrew that never maps anything on the GPU to work
* Stop memory modification check when a invalid address is found
* Fix exception on shader translator when adding goto temp variable
* Support CC on FSET shader instruction, fix CC on ISET.BF shader instruction
* Support texture rectangle targets (non-normalized coords)
* Partial support for shader memory barriers
* Copy 16 bytes at a time for layout conversion, if possible
* Remove some usings that are not needed
* Correct non-constant offset rewrite for texelFetch
* Support non-constant texture offsets on non-NVIDIA gpus
* Use correct shared memory size (in words, not bytes)
* Ensure a valid shared memory size is used
* Use dispatch params shared memory size when available
* Use maximum shared memory size supported by hardware
* Fix draw indexed flag not being cleared for instanced draws, also avoid state updates in the middle of a indexed draw
* Support 3D ASTC textures (using 2D blocks)
* Support bindless texture gather shader instruction
* Support depth clip mode and disable shader fast math optimization on NVIDIA as a workaround for compiler bugs (?)
* Fix wrong maximum id on sampler pool in some cases
* Support shared color mask, implement more shader instructions
* Initial support for the guest OpenGL driver (NVIDIA and Nouveau)
* Add a pass to turn global memory access into storage access, and do all storage related transformations on IR
* Fix texture 0 not being bound for compute
* Partial support for branch with CC, and fix a edge case of branch out of loop on shaders
* Simplified F2I shader instruction codegen
* Implement HSET2 shader instruction and fix errors uncovered by Rodrigo tests
* Fix shader code comparison method
* Optimize RangeList by not doing an allocation on every call to the Find methods
* Add PSET shader instruction
* Revert "Simplify shader uniform buffer access codegen"
* Simplify shader uniform buffer access codegen
* Improved and simplified window texture presentation
* Small optimizations on texture and sampler pool invalidation
* Correct ordering of compute buffer bind
* Share texture pool cache between graphics and compute
* Separate sub-channel state
* Remove TranslatorConfig struct
* Add R2P shader instruction
* Make the shader translator more error resilient (part 2)
* Make the shader translator more error resilient
* Add missing check for BRX, do not add a out of bounds block to the queue
* Invalidate shaders when they are modified
* Add partial support for the BRX shader instruction
* Fix image units bindings and shader dump in the presence of NOPs
* Revert "Check for ASTC support on the list of supported compressed formats"
* Check for ASTC support on the list of supported compressed formats
* Fix regression on load/store local (wrong type)
* Do not throw for invalid ASTC compressed textures
* Add ATOMS, LDS, POPC, RED, STS and VOTE shader instructions, start changing the way how global memory is handled
* Use a more efficient range list on the buffer manager
* Remove some unused constants and other code
* Add partial support for array of samplers, and add pass to identify them from bindless texture accesses
* Support window resizing
* Implement BFI, BRK, FLO, FSWZADD, PBK, SHFL and TXD shader instructions, misc. fixes
* Support copy of slices to 3D textures, remove old 3D render target layered render support, do not delete textures with existing views created from them
* Add shader support for the round mode on the F2F instruction, support mipmaps on ASTC compressed textures
* Fix shader FMUL32I instruction saturation
* Flush buffers on copies
* Add per-source type memory change tracking, simplified state change tracking, other fixes
* Initial support for image stores, support texture sample on compute
* Fix incorrect saturation on HADD2 and HMUL2 shader instructions
* Fix BC6 Signed/Unsigned float texture formats
* Handle quad and quad strip primitive types using triangle fans
* Improved Min/Mip -> MinFilter enum conversion
* Calculate width from stride on texture copies
* Initial work
* Fix timezone month being reported incorrectly to games (#861)
Download: Ryujinx Git (2020/01/10)
Source:Here
Wow
ReplyDelete